Paclitaxel, an anti-microtubule agent extracted from the needles and bark of the Pacific yew tree, Taxus brevifolia, has shown a remarkable anti-neoplastic effect in human cancer in Phase I studies and early Phase II and III trials (Horwitz et at, 1993). This has been reported primarily in advanced ovarian and breast cancer. Significant activity has been documented in small-cell and non-small cell lung cancer, head and neck cancers, and in metastatic melanoma. However, a major difficulty in the development of paclitaxel for clinical trial use has been its insolubility in water.
Docetaxel is semisynthetically produced from 10-deacetyl baccatin III, a noncytotoxic precursor extracted from the needles of Taxus baccata and esterified with a chemically synthesized side chain (Cortes and Pazdur, 1995). Various cancer cell lines, including breast, lung, ovarian, and colorectal cancers and melanomas have been shown to be responsive to docetaxel. In clinical trials, docetaxel has been used to achieve complete or partial responses in breast, ovarian, head and neck cancers, and malignant melanoma.
Paclitaxel is typically formulated as a concentrated solution containing paclitaxel, 6 mg per milliliter of Cremophor EL (polyoxyethylated castor oil) and dehydrated alcohol (50% v/v) and must be further diluted before administration (Goldspiel, 1994). Paclitaxel (Taxol™) has shown significant activity in human cancers, including breast, ovarian, non-small cell lung, and head and neck cancers (Rowinsky and Donehower, 1995). It has also shown significant activity in patients with advanced breast cancer who had been treated with multiple chemotherapeutic agents (Foa et al., 1994). As with most chemotherapeutic agents, however, the maximum tolerated dose of paclitaxel is limited by toxicity. In humans, paclitaxel's major toxic effect at doses of 100–250 mg/m2 is granulocytopenia (Holmes et al., 1995); symptomatic peripheral neuropathy is its principal nonhematologic toxicity (Rowinsky et al., 1993).
The amount of Cremophor EL necessary to deliver the required doses of paclitaxel is significantly higher than that administered with any other drug that is formulated in Cremophor. Several toxic effects have been attributed to Cremophor, including vasodilatation, dyspnea, and hypotension. This vehicle has also been shown to cause serious hypersensitivity in laboratory animals and humans (Weiss et al., 1990). In fact, the maximum dose of paclitaxel that can be administered to mice by i.v. bolus injection is dictated by the acute lethal toxicity of the Cremophor vehicle (Eiseman et al., 1994). In addition, Cremophor EL, a surfactant, is known to leach phthalate plasticizers such as di(2-ethylhexyl)phthalate (DEHP) from the polyvinylchloride bags and intravenous administration tubing. DEHP is known to cause hepatotoxicity in animals and is carcinogenic in rodents. This preparation of paclitaxel is also shown to form particulate matter over time and thus filtration is necessary during administration (Goldspiel, 1994). Therefore, special provisions are necessary for the preparation and administration of paclitaxel solutions to ensure safe drug delivery to patients, and these provisions inevitably lead to higher costs.
Prior attempts to obtain water soluble paclitaxel have included the preparation of prodrugs of paclitaxel by placing solubilizing moieties such as succinate, sulfonic acid, amino acids, and phosphate derivatives at the 2′-hydroxyl group or at the 7-hydroxyl position (Deutsch et al., 1989; Mathew et al., Zhao and Kingston, 1991, 1992; Nicolaou et al., 1993; Vyas et al., 1995, Rose et al., 1997). While some of these prodrugs possess adequate aqueous solubility, few have antitumor activity comparable to that of the parent drug (Deutsch et al., 1989; Mathew et al., 1992; Rose et al., 1997). Several of these derivatives are not suitable for i.v. injection because of their instability in aqueous solution at neutral pH. For example, Deutsch et al. (1989) report a 2′-succinate derivative of paclitaxel, but water solubility of the sodium salt is only about 0.1% and the triethanolamine and N-methylglucamine salts were soluble at only about 1%. In addition, amino acid esters were reported to be unstable. Similar results were reported by Mathew et al. (1992).
Recently, Nicolaou et al. (1993) reported the synthesis and in vitro biological evaluation of a novel type of prodrug termed “protaxols”. These compounds possess greater aqueous solubility and are converted to paclitaxel as the active drug through an intramolecular hydrolysis mechanism. However, no in vivo data on the antitumor activity of protaxols are yet available. Greenwald et al. reported the synthesis of highly water-soluble 2′ and 7-polyethylene glycol esters of paclitaxel (Greenwald et al., 1994). Using the strategy of polymer linkage, others have developed water-soluble polyethylene glycol (PEG)-conjugated paclitaxel (Li et al., 1996; Greenwald et al., 1996). Although these conjugates have excellent water solubility, their therapeutic efficacies are not better than free paclitaxel. Moreover, PEG has only two reactive functional groups at each end of its polymer chain, which effectively limit the amount of paclitaxel that PEG could carry (U.S. Pat. No. 5,362,831).
Other attempts to solve these problems have involved microencapsulation of paclitaxel in both liposomes and nanospheres (Bartoni and Boitard, 1990). The liposome formulation was reported to be as effective as free paclitaxel, however only liposome formulations containing less than 2% paclitaxel were physically stable (Sharma and Straubinger, 1994). Unfortunately, the nanosphere formulation proved to be toxic. There is still a need therefore for a water soluble paclitaxel formulation that can deliver effective amounts of paclitaxel and docetaxel without the disadvantages caused by the insolubility of the drug.
Another obstacle to the widespread use of paclitaxel is the limited resources from which paclitaxel is produced, causing paclitaxel therapy to be expensive. A course of treatment may cost several thousand dollars, for example. There is the added disadvantage that not all tumors respond to paclitaxel therapy, and this may be due to the paclitaxel not getting into the tumor. There is an immediate need, therefore, for effective formulations of paclitaxel and related drugs that are water soluble with long serum half lives for treatment of tumors, autoimmune diseases such as rheumatoid arthritis, as well as for the prevention of restenosis of vessels subject to traumas such as angioplasty and stenting.